Batch 2.2 Analysis - Characterization

In this project sample's from Lisbon are fabricated with large parameter variation. The samples also show variation within pixels. These factors combined make it hard to reach conclusion. We decided to do a large amount of experiments to overcome some of these issues.
To make the most out of this discussion, we made the plots interactive so that each person can explore the data and have better insights.
Parallel coordinates
loSAMPLE SAMPLElo1glo2getchetch1getch2gALDloALDetchTa2O5+ SIo2Die MaTa2O5380Die d100RTDie T200PVDDie PrALDLoSD EtchRISF6NANPass MParyNANPass d[um]6501300
Fab_Batch-2-2
Count: 1
Overview on fabrication methods.
etch2gsamplelo2gALDloetch1getchlo1gloALDetchscanmeasurmentuserMULTIsetupRADRADmfIV-TESTtestOUTIVPULSENAN
Characterization_Batch-2-2_2021-03-09
Overview on performed measurements.
I-V measurements (on 2021-03-01)
−505−11−10−9−8−7−6−5−4
1 iter- 11 iter- 52 iter- 12 iter- 53 iter- 13 iter- 54 iter- 14 iter- 55 iter- 15 iter- 56 iter- 16 iter- 57 iter- 17 iter- 58 iter- 18 iter- 59 iter- 19 iter- 5IV Batch: 2-2 etch- VD: 5.0[V] on date: 2021-03-01VG [V]ID [A]logâ–¼

 etch
Best among non-ALD samples.
Curves have hysteresis, shift towards left after 5 cycles.
Vth ranges [-3V;-1V].
Pixel log10 Leakage [A] On-Off V threshold [V] Mobility [cm^2/(V*s)] Subthreshold slope [V/dec]
1 -9.6 6.55 -1.09 6.17 0.38
2 -9.2 7.67 -1.07 5.80 0.45
3 -9.7 6.52 -1.31 5.26 0.82
4 -9.4 6.15 -1.07 5.60 0.48
5 -9.6 6.36 -1.11 5.33 0.52
6 -9.7 4.46 -1.49 4.09 1.26
7 -9.9 3.30 -0.95 5.31 1.20
8 -9.2 3.44 -0.44 4.80 1.12
9 -9.6 6.62 -1.16 5.32 0.45
−505−11−10−9−8−7−6−5−4
2 iter- 12 iter- 53 iter- 13 iter- 54 iter- 14 iter- 56 iter- 16 iter- 57 iter- 17 iter- 58 iter- 18 iter- 59 iter- 19 iter- 5IV Batch: 2-2 lo- VD: 4.99999[V] on date: 2021-03-01VG [V]ID [A]logâ–¼

 lo
Curves have hysteresis, shift towards left after 5 cycles.
Vth ranges [-3V;-1V].
Pixel log10 Leakage [A] On-Off V threshold [V] Mobility [cm^2/(V*s)] Subthreshold slope [V/dec]
2 -9.0 6.61 -2.58 4.81 0.26
3 -10.1 7.62 -1.45 4.92 0.18
4 -10.0 6.88 -1.69 5.99 0.25
6 -10.0 6.21 -2.03 1.70 0.50
7 -9.9 2.45 -1.54 5.62 1.72
8 -9.9 2.29 -2.08 3.64 1.73
9 -10.1 7.55 -0.97 4.52 0.22
−505−12−11−10−9−8−7−6−5−4
1 iter- 11 iter- 52 iter- 12 iter- 53 iter- 13 iter- 54 iter- 14 iter- 55 iter- 15 iter- 56 iter- 16 iter- 59 iter- 19 iter- 5IV Batch: 2-2 etch1g- VD: 5.0[V] on date: 2021-03-01VG [V]ID [A]logâ–¼

 etch1g
Curves have hysteresis, shift towards left after 5 cycles.
Vth ranges [-3V;-1V].
Pixel log10 Leakage [A] On-Off V threshold [V] Mobility [cm^2/(V*s)] Subthreshold slope [V/dec]
1 -10.2 6.51 -2.20 3.83 0.21
2 -9.8 6.55 -1.79 6.48 0.55
3 -5.6 5.25 -1.83 6.25 0.69
4 -10.3 6.46 -1.28 4.78 0.58
5 -10.2 6.42 -1.71 4.28 0.63
6 -9.4 6.60 -1.98 6.18 0.47
9 -10.3 6.65 -1.81 5.69 0.50
−505−12−11−10−9−8−7−6−5−4
3 iter- 13 iter- 54 iter- 14 iter- 57 iter- 17 iter- 59 iter- 19 iter- 5IV Batch: 2-2 lo1g- VD: 5.0[V] on date: 2021-03-01VG [V]ID [A]logâ–¼

 lo1g
Only 3 out of 9 working.
Curves have hysteresis, shift towards left after 5 cycles.
Pixel log10 Leakage [A] On-Off V threshold [V] Mobility [cm^2/(V*s)] Subthreshold slope [V/dec]
3 -9.4 7.50 -0.40 4.44 0.15
4 -9.9 6.96 -1.30 5.14 0.32
7 -10.1 3.98 -1.83 4.31 1.27
9 -10.3 6.61 -1.04 4.79 0.37
−505−11−10−9−8−7−6−5−4
2 iter- 12 iter- 53 iter- 13 iter- 54 iter- 14 iter- 55 iter- 15 iter- 56 iter- 16 iter- 58 iter- 18 iter- 59 iter- 19 iter- 5IV Batch: 2-2 etch2g- VD: 4.99999[V] on date: 2021-03-01VG [V]ID [A]logâ–¼

 etch2g
Curves have hysteresis (stronger than lo2g), shift towards left after 5 cycles.
Vth ranges [-3V;-1V].
Pixel log10 Leakage [A] On-Off V threshold [V] Mobility [cm^2/(V*s)] Subthreshold slope [V/dec]
2 -9.9 7.19 -1.48 5.90 0.37
3 -9.6 6.54 -0.64 5.71 0.46
4 -10.3 6.60 -1.77 5.62 0.39
5 -10.2 6.38 -3.04 2.74 0.48
6 -9.9 7.67 -1.39 5.41 0.41
8 -10.2 6.86 -1.55 5.04 0.42
9 -10.3 6.97 -1.44 5.22 0.18
−505−11−10−9−8−7−6−5−4
1 iter- 11 iter- 52 iter- 12 iter- 53 iter- 13 iter- 54 iter- 14 iter- 55 iter- 15 iter- 57 iter- 17 iter- 58 iter- 18 iter- 5IV Batch: 2-2 lo2g- VD: 4.99999[V] on date: 2021-03-01VG [V]ID [A]logâ–¼

 lo2g
Curves have hysteresis, shift towards left after 5 cycles.
Vth ranges [-5V;-1V].
Pixel log10 Leakage [A] On-Off V threshold [V] Mobility [cm^2/(V*s)] Subthreshold slope [V/dec]
1 -9.9 6.71 -4.35 4.83 0.48
2 -10.3 7.63 -1.64 4.64 0.91
3 -10.3 6.58 -1.70 4.27 0.42
4 -10.1 6.54 -2.25 3.79 0.45
5 -10.1 5.65 -3.02 3.57 1.13
7 -10.0 7.66 -1.68 5.05 0.43
8 -9.8 6.66 -2.43 5.01 0.33
−2024−9−8−7−6−5−4
2 iter- 12 iter- 53 iter- 13 iter- 55 iter- 15 iter- 56 iter- 16 iter- 57 iter- 17 iter- 58 iter- 18 iter- 59 iter- 19 iter- 5IV Batch: 2-2 ALDetch- VD: 5.0[V] on date: 2021-03-01VG [V]ID [A]logâ–¼

 ALDetch
Less reproducible than ALDlo, devices 2 (broken, high leakage) and 3 differ from other samples.
Vth around -2V, stable after 5 measurement cycles.
I_off higher than other samples (stronger leakage).
Pixel log10 Leakage [A] On-Off V threshold [V] Mobility [cm^2/(V*s)] Subthreshold slope [V/dec]
2 -1.5 2.64 -1.58 1.66 0.56
3 -9.4 5.13 -0.66 4.75 0.27
5 -9.7 3.28 -1.85 7.90 0.23
6 -10.2 2.70 -2.05 8.31 0.31
7 -10.0 4.25 -1.50 8.11 0.27
8 -10.1 3.53 -1.73 8.20 0.32
9 -10.0 2.79 -2.00 8.35 0.31
−4−2024−9−8−7−6−5−4
1 iter- 11 iter- 53 iter- 13 iter- 54 iter- 14 iter- 55 iter- 15 iter- 56 iter- 16 iter- 57 iter- 17 iter- 58 iter- 18 iter- 59 iter- 19 iter- 5IV Batch: 2-2 ALDlo- VD: 4.99999[V] on date: 2021-03-01VG [V]ID [A]logâ–¼

 ALDlo
Highest reproducibility.
Vth approximaltely = 0V, stable after 5 measurement cycles.
I_off higher than other samples (stronger leakage).
Pixel log10 Leakage [A] On-Off V threshold [V] Mobility [cm^2/(V*s)] Subthreshold slope [V/dec]
1 -10.3 5.31 -0.38 6.85 0.17
3 -6.3 5.17 -0.37 7.62 0.06
4 -10.5 5.37 -0.19 6.93 0.18
5 -10.3 5.30 -0.42 7.80 0.19
6 -10.4 5.02 -0.14 7.82 0.23
7 -10.4 5.28 -0.34 7.40 0.20
8 -10.3 5.04 -0.31 7.90 0.20
9 -10.4 5.31 -0.31 8.44 0.20
Form a charcterization perspective, we see ALDlo as the new benchmark.
Output measurements (on 2021-03-01)
01234505μ10μ15μ20μ
0.00.30.60.91.21.51.82.12.42.73.03.33.63.9OUT Batch: 2-2 lo on date: 2021-03-01VD [V]ID [A]pixel 2â–¼

lo
01234502μ4μ6μ8μ10μ12μ14μ
0.00.30.60.91.21.51.82.12.42.73.03.33.63.9OUT Batch: 2-2 etch on date: 2021-03-01VD [V]ID [A]pixel 1â–¼

etch
01234501μ2μ3μ4μ5μ6μ
0.00.30.60.91.21.51.82.12.42.73.03.33.63.9OUT Batch: 2-2 lo1g on date: 2021-03-01VD [V]ID [A]pixel 3â–¼

lo1g
01234502μ4μ6μ8μ10μ12μ14μ
0.00.30.60.91.21.51.82.12.42.73.03.33.63.9OUT Batch: 2-2 etch1g on date: 2021-03-01VD [V]ID [A]pixel 1â–¼

etch1g
01234505μ10μ15μ20μ25μ30μ35μ
-0.00.30.60.91.21.51.82.12.42.73.03.33.63.9OUT Batch: 2-2 lo2g on date: 2021-03-01VD [V]ID [A]pixel 1â–¼

lo2g
01234502μ4μ6μ8μ10μ12μ14μ16μ
-0.00.30.60.91.21.51.82.12.42.73.03.33.63.9OUT Batch: 2-2 etch2g on date: 2021-03-01VD [V]ID [A]pixel 2â–¼

etch2g
012345020μ40μ60μ80μ100μ120μ140μ160μ180μ
-0.00.30.60.91.21.51.82.12.42.73.03.33.63.9OUT Batch: 2-2 ALDlo on date: 2021-03-01VD [V]ID [A]pixel 1â–¼

ALDlo
012345010μ20μ30μ40μ50μ60μ70μ80μ
-0.00.30.60.91.21.51.82.12.42.73.03.33.63.9OUT Batch: 2-2 ALDetch on date: 2021-03-01VD [V]ID [A]pixel 2â–¼

ALDetch

Scatter matrix plot.

Summary of characterization results.